Candida albicans strain B 311 (ATCC 32354) was grown aerobically overnight in Sabouraud’s dextrose broth (Thermo Fisher Scientific, Houston, TX) at 37˚C. Lactobacillus rhamnosus GR-1® and Lactobacillus reuteri RC-14®, which were investigated because they are used in commercial vaginal probiotic products, (provided by Dr. Gregor Reid, Urex Bioscience, London, Ontario) were grown in anaerobic MRS broth (Thermo Fisher Scientific) at 35˚C.

The vaginal epithelial cell line VK2 E6/E7 (CRL-2616, American Type Culture Collection, Rockville, MD.) was seeded at 5 × 10⁵ cells/ml and grown to confluence on 35 mm polyester inserts [14,15] in Keratinocyte-Serum Free medium (K-SFM) containing 5 ng/ml recombinant epidermal growth factor and 50 μg/ml bovine pituitary extract (Invitrogen Corporation, Grand Island, NY) and primed with a physiological 10 nM concentration of 17 β-estradiol in the basolateral medium [16]. Genital tract secretions (GTS) medium [17] was used on the apical side of the VK2 cell monolayers. The cells were grown at 37˚C with a 5% CO₂ atmosphere and 100% humidity. Confluence, polarization, and differentiation of the VK2 cell monolayers were assessed by measurement of TEER with an EVOM device (World Precision Instruments, Inc., Sarasota, FL). Probiotic lactobacilli were added at 1 × 10⁷ CFU/ml to the apical culture chambers 4 hours before 18 hour challenge with 2 × 10⁶ CFU/ml C. albicans. The experimental groups compared in this study were: 1) VK2 cells grown 18 hours with K-SFM medium on the basolateral sides and GTS medium on the apical sides of the culture inserts; 2) VK2 cells cultured in the same media and challenged 18 hours on the apical side with 2 × 10⁶ CFU C. albicans; 3) VK2 cells with 10 nM 17β-estradiol added to the basolateral medium; 4) VK2 cells with 10 nM 17β-estradiol added to the basolateral medium and challenged on the apical side with 2 × 10⁶ CFU C. albicans; 5) VK2 cells cultured with 1 × 10⁷ CFU of both Lactobacillus strains on the apical side, and 6) VK2 cells cultured with 10 nM 17β-estradiol added to the basolateral medium and pre-incubated 4 hours with 1 × 10⁷ CFU of both Lactobacillus strains on the apical side before apical challenge with 2 × 10⁶ CFU C. albicans. The final incubations were 18 hours after C. albicans challenge.

Differences in nutritional content of K-SFM and GTS to be used on the basolateral and apical sides, respectively, of the cell monolayers prompted the need to evaluate growth and stability of the VK2 cells and microbes in each medium. A 24-well plate of VK2 cells seeded at 5 × 10⁵ cells/ml on polycarbonate inserts in complete K-SFM was grown to confluence and maximal TEER at 7 days after plating. The K-SFM apical chamber media in 6 wells was replaced with GTS and grown with media changes every two days. Six wells were set up with GTS containing 10 nM 17β-estradiol; another 6 wells were set up with K-SFM and 17β-estradiol. The monolayers were observed and TEER was measured daily. After a week, the cells were harvested from the inserts with trypsin/EDTA (Invitrogen) treatment and the numbers of viable and non-viable cells were counted in a hemacytometer using trypan blue stain. The effects of VK2 cell co-culture with microbes were assessed in the same manner.

50 µl aliquots of an 18-hour C. albicans culture in Sabouraud’s dextrose broth was diluted into 250 µl of GTS medium containing serial dilutions of H₂O₂ or lactic acid (Sigma Chemical Co., St. Louis, MO) into wells of a 100-well honeycomb plate (Growth Curves, USA, Haverhill, MA) sealed with vinyl tape. The plates were incubated with shaking at 300 RPM and at a temperature of 37˚C in a Bioscreen C instrument (Growth Curves) that recorded optical densities of growth in the wells at 15 minutes intervals for 18 hours using a broad band filter at wavelengths of 420 - 580 nm. Growth curves were generated and growth rates were estimated with Norden Lab v. 7.2 software (Norden Logic Oy, Helsinki, Finland) and Prism v. 5.0 software (GraphPad Software, San Diego, CA).

50 µl aliquots of an 18-hour C. albicans culture in Sabouraud’s dextrose broth were diluted into 250 µl of GTS medium containing 10 nM 17β-estradiol or 10 mM ICI_182,780, or containing both compounds in wells of a 96-well assay plate. The same treatments were also evaluated in a honeycomb plate for growth curve analyses. There were 6 replicates per treatment group. The assay plates were incubated 18 hours in a 37˚C incubator. The cultures were diluted 10-fold in Dulbecco’s phosphate-buffered saline (Invitrogen) and 100 µl samples were spread onto microscope slides with a Cytofuge 2 cytocentrifuge for 4 minutes at 2200 RPM (Statspin, Norwood, MA). The slides were stained with Gram’s stain and fungal blastospores and hyphae were counted and averaged from 10 high power (400×) microscopic fields.

Gram stains were also analyzed to compare inhibitory effects of 0.01 µg/ml H₂O₂, 2 µg/ml lactic acid, 1 × 10⁷ CFU/ml L. rhamnosus GR-1 or L. reuteri RC-14, or a combination of 1 × 10⁷ CFU/ml of both strains of lactobacilli. The results were compared as ratios of hyphae to blastospores. The % hyphae present in microscopic fields were determined as numbers of hyphae counted/numbers of hyphae + numbers of blastospores × 100.

A semiquantitative assay for adherence of lactobacilli to the VK2 cells was employed [7]. Confluent cultures of VK2 cells on polyester inserts (Thermo Fisher) were co-cultured with lactobacilli at a concentration of 1 × 10⁷ CFU/ml, and challenged with 2 × 10⁶ CFU/ml of the virulent C. albicans strain, as described [7]. After 18 hours incubation at 37˚C in a 5% CO₂ atmosphere with 100% humidity, the culture inserts were rinsed four times with GTS medium and the polyester membranes with the VK2 cell monolayers were removed from the inserts. The membranes were fixed with 3.7% paraformaldehyde for 1 hour. The membranes were rinsed with GTS medium and the cells were stained with Gram’s crystal violet solution. The average number of bacteria and yeast cells adhering to VK2 cells were counted in 10 random microscopic fields.

The ability of the probiotic bacteria to inhibit growth of C. albicans B 311 was evaluated with a culture overlay technique [7]. Each Lactobacillus strain was diluted in MRS broth (Thermo Fisher) to 9 × 10⁸ CFU/ml and spread in 20 mm bands across MRS agar plates and then grown 24 hours at 35˚C in an anaerobic chamber (Coy Laboratories, Grass Lake, MI). The cultures were overlaid with melted YM agar (Thermo Fisher). The plates were swabbed with C. albicans B 311 in Sabouraud broth diluted to 1.5 × 10⁸ CFU/ml, pre-incubated at 4˚C for 4 hours to harden the agar, and then incubated overnight at 37˚C aerobically. Inhibition of C. albicans growth was observed visually and scored semi-quantitatively as: (−) no inhibition, (+) partial growth inhibition over the lactobacilli, (++) complete inhibition over the lactobacilli, or (+++) complete inhibition over and beyond the band of lactobacilli [7].

The amount of H₂O₂ produced by the lactobacilli in culture was measured with an iodometric method designed for determinations from bacterial cultures [18]. Lactobacillus spp. cultures (20 ml) were grown in Rogosa broth (Thermo Fisher) 48 hours at 35˚C anaerobically and centrifuged at 1800 × g for 15 minutes. Technical duplicates of each control and blank (Rogosa broth) and 6 replicates from each Lactobacillus sp. culture supernatant were added (50 µl) to 2 ml of HCl, 0.2 ml of KI, 0.2 ml of 1 mm ammonium molybdate solution (Sigma), and 0.2 ml of starch solution (Sigma), as described [18]. The optical absorbance was measured with a spectrophotometer at 570 nm wavelength and the standard curve from serial dilutions of H₂O₂ and concentrations of samples were determined.

The amounts of H₂O₂ produced by the lactobacilli applied to the apical sides of VK2 cell monolayers were measured with a chromagenic method designed to measure H₂O₂ concentrations in vaginal secretions [18]. The GTS medium from the apical chambers and K-SFM medium from the basolateral chambers conditioned by the lactobacilli were diluted in 0.01 M sodium acetate, pH 4.5, mixed with 0.01 mg/ml peroxidase (Sigma) and 20 ml o-dianisidine solution (Sigma), and incubated at room temperature 10 minutes. The reactions were stopped with 40 ml of 4N HCl and incubated 5 minutes at room temperature. Color development was measured in a plate reader (Perkin Elmer, Waltham, MA) at 405 nm. Concentrations were determined from a standard curve consisting of dilutions of a 30% H₂O₂ stock solution (Sigma).

The PHR-146 Micro Combination pH Electrode from Lazar Research (Thermo Fisher) was used to measure pH levels in cell-conditioned media. After the 18 hour incubation of VK2 cells with the microbes, media was removed and saved from the inserts and the wells; both were centrifuged at 10,000 × g for 2 minutes, and the pH of each supernatant was measured using the micro pH electrode.

Normally distributed results were analyzed for statistical significance by paired T-tests where appropriate and by Repeated Measures One-Way Analysis of Variance and Newman-Keuls post tests for multiple comparisons [19] with Prism v. 5.0 software (GraphPad Software). Statistical significance was defined by a P < 0.05.

The environment surrounding vaginal epithelial cells in vivo is different on the apical and basolateral sides of the vaginal mucosa. The effect of the media surrounding the VK2 cells grown on culture inserts on monolayer stability was assessed. Over seven days of culture, confluent monolayers had increased TEER values from 32 to 48 Ω cm². No differences in trypan blue uptake or cell counts were observed. On the fourth and seventh days of culture, the TEER values in K-SFM were higher than in GTS.

When the fungi and bacteria were inoculated into GTS individually to test for differences in growth characteristics, the numbers of C. albicans cells increased from 2 × 10⁶ CFU/ml to 5.3 × 10⁷ CFU/ml overnight. Addition of 17β-estradiol to the medium increased the average number of C. albicans to 1.2 × 10⁸ CFU/ml. Growth curves of C. albicans in GTS had significantly increased slopes from 12.42 ± 0.57 to 17.96 ± 0.87 (average R² = 0.948 and 0.945, respectively). Slopes were determined by regression analysis with Prism software, over 18 hours’ incubation. The lactobacilli did not increase in numbers overnight in GTS or GTS containing 17β-estradiol; however, they remained viable in GTS medium for 48 hours, since growth resumed after transfer to MRS medium.

The presence of C. albicans appeared to increase the TEER of the VK2 cell monolayers of most samples, but the effect was not statistically significant (Figure 1). When 17b-estradiol was incubated with the cells alone, there was a significant increase in TEER, but when VK2 cells were treated with 17b-estradiol and C. albicanschallenged, the TEER values were significantly reduced (Figure 1). The probiotic lactobacilli did not significantly alter the TEER values of VK2 cells alone or modify the reduced TEER values of VK2 cells treated with 17b-estradiol and challenged with C. albicans (Figure 1).

Assays of probiotic mechanisms were performed to assess the ability of L. rhamnosus GR-1 and L. reuteri RC-14 to inhibit the growth and infectivity of the C. albicans B-311 strain. The fungal inhibition assay revealed that both Lactobacillus spp. strains were inhibitory to C. albicans growth, which was inhibited 2+ by L. rhamnosus GR-1 and 3+ by L. reuteri RC-14 (data not shown) on the 4-point scale of Strus, et al. [7]. This corresponds to L. reuteri RC-14 producing more soluble factors that diffuse further into the agar overlay to inhibit C. albicans growth than those produced by L. rhamnosus GR-1.

As assessed in this study, the production of H₂O₂ by lactobacilli is an inhibitory mechanism of fungal growth. The amount of H₂O₂ produced by L. rhamnosus GR-1 and L. reuteri RC-14 in culture supernatants averaged

12.5 µg/ml. These H₂O₂ concentrations were 20-fold greater than H₂O₂ concentrations present in the apical GTS media from the VK2 cell cultures containing lactobacilli. This difference in H₂O₂ concentrations produced by the lactobacilli in anaerobic enriched media compared with an environment approximating the vaginal mucosa suggests that the Lactobacillus strains tested may not be as adaptable to that environment as normal vaginal isolates. The H₂O₂ concentrations in VK2 cell cultures with or without 17β-estradiol, and VK2 cell cultures challenged with C. albicans were not different (Figure 2A). Addition of 17β-estradiol to the C. albicans-challenged VK2 cell cultures yielded significantly increased H₂O₂ concentrations in the basolateral medium (Figure 2A).

The probiotic lactobacilli did not induce H₂O₂ production by VK2 cells, and did not appear to contribute significantly to H₂O₂ in the co-culture with VK2 cells and C. albicans (Figure 2A).

Production of H₂O₂ by VK2 cells was not expected, but the cells are known to produce soluble factors that inhibit Candida spp. [8]. The H₂O₂ concentration was also significantly elevated in the basolateral media from the C. albicans-challenged group containing 17β-estradiol and the probiotic lactobacilli.

The average pH of VK2 cell cultures was not significantly affected by overnight challenge with C. albicans or 17β-estradiol. Co-culture of VK2 cells with the lactobacilli significantly decreased the pH, in both the apical and basolateral media (Figure 2B). The presence of C. albicans and 17β-estradiol modulated the low pH produced by the probiotic lactobacilli.

C. albicans adhered to VK2 cell monolayers as hyphae and blastospores (Figure 3A). The lactobacilli also appeared to adhere to the VK2 cells, and in their presence, C. albicans remained more in the blastospore form (Figure 3B). The numbers of adherent C. albicans on VK2 cells were significantly decreased in the presence of lactobacilli (Figure 3C).

Reduced adherence of C. albicans to VK2 cells in the presence of probiotic lactobacilli may be related to decreased pH resulting from acid production by the lactobacilli. This conclusion agrees with the observation by Gaffney et al., that lowered pH of vaginal fluids corresponds with reduced microbial adhesion to epithelial cell lines [20].

Growth inhibitory concentrations of H₂O₂ and lactic acid for C. albicans, L. rhamnosus, and L. reuteri were determined using growth curve analyses. The inhibitory concentration of H₂O₂ on C. albicans was 0.3 µg/ml (Figure 4A). This was also the concentration maximally achieved by the lactobacilli co-cultured with VK2 cells. The inhibitory concentration of lactic acid on C. albicans growth was 10 µg/ml (Figure 4B).

Growth curves of L. rhamnosus GR-1 showed inhibitory concentrations of 50 µg/ml H₂O₂ and 6 mg/ml lactic acid (Figure 5). The growth curves of L. reuteri RC-14 showed inhibited growth at 2 mg/ml H₂O₂ or 6 mg/ml lactic acid over an 18-hour time course (Figure 5).

Addition of 10 nM 17β-estradiol increased C. albicans resistance to growth inhibition by 0.3 mg/ml H₂O₂, but not 10 mg/ml lactic acid (Figure 6).

Addition of 10 nM 17β-estradiol to C. albicans cultures

was associated with an apparent increase in numbers of hyphal forms of the fungus (Figure 7). The percentages of hyphae in 10 microscopic fields per control or treatment increased significantly from 22% to 32%, respectively, when 10 nM 17β-estradiol was present during incubation.

The ability of 17β-estradiol to induce hyphal germination may be counteracted by cytotoxic molecules such as H₂O₂ and lactic acid. The probiotic lactobacilli may also have influenced 17β-estradiol induction of hyphal germination. To assess this, C. albicans was grown in GTS medium with or without 10 nM 17β-estradiol and treated with 0.01 mg/ml H₂O₂, 2 mg/ml lactic acid, or with 1 × 10⁷ CFU/ml L. rhamnosus GR-1, L. reuteri RC-14, or both strains together. The ratios of hyphae to blastospores in the cultures with or without 17β-estradiol were decreased in the presence of H₂O₂ or lactic acid as compared to the control group, but neither compound was able to prevent stimulation of hyphal germination by 17β-estradiol (Figure 8). Figure 8 also shows that the presence of L. rhamnosus GR-1 did not significantly influence the ratios of hyphae and blastospores nor was it able to prevent stimulation of hyphal germination by 17β-estradiol. The presence of L. reuteri RC-14 significantly reduced the hyphae to blastospore ratio compared with the control containing 17β-estradiol, but not as compared with the untreated control. The presence of L. reuteri RC-14, with or without L. rhamnosus GR-1, significantly reduced the ratio of hyphae to blastospores in the 17β-estradiol-treated group compared with the control group (Figure 8). These results suggest that L. reuteri RC-14, but not L. rhamnosus GR-1, inhibits 17β-estradiol-induced hyphal germination.